Ack, code review only.
Thanks
Praveen
On 01-Sep-17 9:24 AM, Gary Lee wrote:
---
src/amf/amfd/sg.cc | 5 +
src/amf/amfd/su.cc | 6 ++
2 files changed, 11 insertions(+)
diff --git a/src/amf/amfd/sg.cc b/src/amf/amfd/sg.cc
index 8f3590190..986bb 100644
--- a/src/amf/amfd/sg.cc
+++
I got confused with previous release strategies.
Please push in release branch also.
Thanks
Praveen
On 31-Aug-17 11:48 AM, praveen malviya wrote:
I am not seeing any milestone for any maintenance/release branch.
There is only one milestone 5.17.10 which is for new release.
Thanks
I am not seeing any milestone for any maintenance/release branch.
There is only one milestone 5.17.10 which is for new release.
Thanks
Praveen
On 31-Aug-17 11:34 AM, Gary Lee wrote:
OK - thanks.
Should we push it to release as well?
On 31/08/17 15:30, praveen malviya wrote:
Hi Gary,
ack
Hi Gary,
ack for part2.
I think I did not grep on saAmfSGNumPrefInserviceSUs in all files.
I have pushed part1. Please push part2.
Thanks
Praveen
On 31-Aug-17 9:50 AM, Gary Lee wrote:
Hi Praveen
ack, but I think further changes are required (see attachment).
Do you think you could push it
Ack, code review only.
Thanks,
Praveen
On 14-Aug-17 9:05 AM, Gary Lee wrote:
CLM and MDS callbacks are delivered to the main thread via different paths.
If a node is restarted quickly, sometimes CLM JOIN is processed before the
prior MDS down. This means the node will not be able to join the c
, "praveen malviya" wrote:
Hi Gary,
Please find one comment inline with [Praveen].
Thanks
Praveen
On 05-Jul-17 2:15 PM, Gary Lee wrote:
> ---
> src/amf/amfnd/avnd_cb.h | 3 +-
> src/amf/amfnd/avnd_di.h | 36
Hi Gary,
Please find one comment inline with [Praveen].
Thanks
Praveen
On 05-Jul-17 2:15 PM, Gary Lee wrote:
---
src/amf/amfnd/avnd_cb.h | 3 +-
src/amf/amfnd/avnd_di.h | 36 --
src/amf/amfnd/avnd_mds.h | 11 +-
src/amf/amfnd/di.cc | 321 +---
Hi Gary,
I have started reviewing it.
Thanks
Praveen
On 15-Aug-17 7:02 AM, Gary Lee wrote:
Hi
I would like to push this in a week's time if no one has comments.
Thanks
Gary
On 05/07/17 18:45, Gary Lee wrote:
---
src/amf/amfnd/avnd_cb.h | 3 +-
src/amf/amfnd/avnd_di.h | 36 --
Ack.
Thanks
Praveen
On 16-Aug-17 1:56 PM, Nguyen Luu wrote:
Summary: amf: update PR doc compliance report for
saAmfComponentErrorClear_4() [#2540]
Review request for Trac Ticket(s): #2540
Peer Reviewer(s): AMF devs
Pull request to: AMF maintainers
Affected branch(es): default
Development branc
Ack, code review only.
Thanks
Praveen
On 16-Aug-17 1:44 PM, Nguyen Luu wrote:
When called with a non-existing component name, saAmfComponentErrorClear_4
should return SA_AIS_ERR_NOT_EXIST instead of SA_AIS_ERR_BAD_OPERATION
as previously done.
---
src/amf/amfnd/err.cc | 17 +++--
Hi,
Both ErrorReport() and ErrorClear() APIs can be called for any component
hosted anywhere.
Since amfnd only maintains local components, component may not be found
in its data base. A message should be sent to AMFD or AMFND should
instantly read IMM database for validating the component.
I
Ack, code review only.
Thanks
Praveen
On 02-Aug-17 9:31 AM, Nguyen Luu wrote:
When called with an uninitialized or already finalized handle,
saAmfPmStart_3 and saAmfResponse_4 should return SA_AIS_ERR_BAD_HANDLE
instead of SA_AIS_ERR_VERSION as previously done.
---
src/amf/agent/amf_agent.cc
Ack.
Thanks
Praveen
On 01-Aug-17 4:42 PM, Anders Widell wrote:
Provide the node address as a command-line parameter when calling the scale-out
script. This can be useful if the scale-out script needs to contact the node
(e.g. copy some files to it or update some configuration on the node's loc
to select the internal OpenSAF
communication address or some other address to present in
saClmNodeCurrAddress.
regards,
Anders Widell
On 08/07/2017 07:12 AM, praveen malviya wrote:
Hi Anders,
I have started reviewing this patch.
One initial query:
We have two sets of attributes for
Hi Anders,
I have started reviewing this patch.
One initial query:
We have two sets of attributes for address in "SaClmNode":
set A) saClmNodeAddressFamily & saClmNodeAddress and
set B )saClmNodeCurrAddressFamily & saClmNodeCurrAddress.
For ticket #2479, its description says making set B as Mu
Hi Anders,
I will review both #2538 and #2538 by Monday.
Thanks,
Praveen
On 04-Aug-17 12:24 PM, Anders Widell wrote:
Hi!
Did you get a chance to look at this yet?
thanks,
Anders Widell
On 07/31/2017 03:10 PM, Anders Widell wrote:
Summary: clm: Include boot time and node address in join r
since after SC absence stage the SG
operation resumes to what it was before loss of SCs.
Thanks,
Minh
On 02/08/17 21:18, praveen malviya wrote:
Hi Minh,
I wanted to highlight a valid case when quiesced HA state can be there
in two SUs in 2N model. In switchover situation when one SU
AMFD sends it a quiesced HA state. Thus there can be two
quiesced valid SUSI in a SG at momentarily.
Thanks,
Praveen
On 02-Aug-17 4:41 PM, minh chau wrote:
Hi Praveen,
Please find my reply inline.
Thanks,
Minh
On 02/08/17 20:17, praveen malviya wrote:
Hi Minh,
Please find one query inline
Hi Minh,
Please find one query inline with [Praveen].
Thanks,
Praveen
On 28-Jul-17 7:44 AM, Minh Chau wrote:
Symtomp is similar to #2477, this patch fixes for case of
2 STANDBY assignment for same SI
---
src/amf/amfd/si.cc| 14 ++
src/amf/amfd/si.h | 1 +
src/amf/amfd/si
Ack for all the patches.
Thanks,
Praveen
On 24-Jul-17 7:27 PM, Anders Widell wrote:
Avoid spamming the syslog with more than one log message in case CLM returns
SA_AIS_ERR_UNAVAILABLE (i.e. we are running on a currently unconfigured node).
---
src/amf/amfd/clm.cc | 6 --
src/amf/amfnd/
I am reviewing the patches.
Thanks,
Praveen
On 24-Jul-17 7:27 PM, Anders Widell wrote:
Summary: clm: Make it possible for a node to scale out itself using autoscaling
[#2509]
Review request for Ticket(s): 2509
Peer Reviewer(s): Praveen
Pull request to:
Affected branch(es): develop
Development
Hi Anders,
Attached is the patch after incorporating the comments.
I will be pushing it on Monday.
Please go through it.
Thanks
Praveen
On 18-Jul-17 9:20 PM, Anders Widell wrote:
Ack with comments:
* Indentation seems to be according to Google C++ style guide, although
the file is written in
Ack.
Thanks
Praveen
On 21-Jul-17 9:52 AM, Gary Lee wrote:
---
src/clm/apitest/tet_saClmClusterTrack.c | 30 +++---
1 file changed, 15 insertions(+), 15 deletions(-)
diff --git a/src/clm/apitest/tet_saClmClusterTrack.c
b/src/clm/apitest/tet_saClmClusterTrack.c
index
Ack, code review only.
Thanks,
Praveen
On 14-Jul-17 6:42 PM, Zoran Milinkovic wrote:
CLM handles ERR_BAD_HANDLE for saImmOmSearchInitialize in
clms_cluster_config_get.
As part of this patch, handling of IMM version is improved in the same function.
---
src/clm/clmd/clms_imm.c | 18 ++
Ack.
Thanks
Praveen
On 05-Jul-17 8:23 PM, Zoran Milinkovic wrote:
The patch removes dependencies between CLM tests. CLM tests can be run more
times now.
Duplicated CLM tests are removed from clmtest.
---
src/clm/apitest/clmtest.c | 15
src/clm/apitest/tet_saClmClusterTra
ess_event: evt->rcv_evt 8
Jun 20 17:20:45.587742 osafamfd [10514:10514:src/amf/amfd/main.cc:0818]
<< process_event
Jun 20 17:20:46.048978 osafamfd [10514:10514:src/amf/amfd/main.cc:0770]
>> process_event: evt->rcv_evt 4
Jun 20 17:20:46.052609 osafamfd [10514:10514:src/amf/amfd/m
Hi Gary,
When SC joins back after SC Absence state, there will not be many
messages in the queue to be processed even in large cluster also.
Also there is no error like ipc send failure.
Is AMFD stuck somewhere and it could not take any message from queue?
I think raising priority will not solv
Ack.
Thanks,
Praveen
On 05-Jul-17 2:49 PM, Minh Chau wrote:
ntfd can sometimes receive SIG_TERM in stopping opensafd even surveillance
thread and ntfimcnd have not been started.
The patch prevents to stop surveillance thread and ntfimcnd in such case.
---
src/ntf/ntfd/ntfs_imcnutil.c | 13 +++
Ack, code review only.
Thanks
Praveen
On 03-Jul-17 6:04 AM, minh chau wrote:
Hi Lennart, Praveen
If we don't have any comments, I would like to push the patch today
Thanks,
Minh
On 28/06/17 14:32, Minh Chau wrote:
In the scenario of shutting down SC while SC switchover is on going,
ntfd co
Hi Gary,
Thanks for the reviewing the patches.
Please see inline with [Praveen].
Thanks,
Praveen
On 28-Jun-17 12:56 PM, Gary Lee wrote:
I think currently the callback is called in the context of the agent MDS
thread? This could introduce threading issues to an application as they
normally e
Hi Lennart,
NTFIMCN initializes with IMM with SAF version A.02.12.
IMM integration with CLM is done in last release and for that SAF
version is A.02.18 (src/imm/README mentions it).
Like any legacy application, NTFIMCN should not get ERR_UNAVAILABLE from
IMM for any API call.
Thanks,
Praveen
create the object. This patch doesn't mean to change any logic, it
just doesn't make the log confusing
Thanks,
Minh
On 27/06/17 18:33, praveen malviya wrote:
Hi Minh,
One comment inline with [ Praveen].
Thanks
Praveen
On 25-May-17 12:53 PM, Minh Chau wrote:
---
src/amf/amfd/
Ack.
Thanks,
Praveen
On 27-Jun-17 3:09 PM, Hoang Vo wrote:
clmtest 7 7 check saClmClusterNodeGet with non-member node,
previously designed as 0x2060F.
change test node_id to 0x2990F to avoid future conflict
when test environment is upgraded.
---
src/clm/apitest/tet_saClmClusterNodeGet.c | 4 +
Ack, code review only.
Thanks
Praveen
On 28-Jun-17 6:48 AM, Gary Lee wrote:
Hi
I would like to push this patch on Friday if there are no objections.
For the medium term, we should consider making AMFND an applier to avoid these
issues?
Thanks
Gary
On 15/6/17, 5:41 pm, "Gary Lee" wrote:
Hi Minh,
One comment inline with [ Praveen].
Thanks
Praveen
On 25-May-17 12:53 PM, Minh Chau wrote:
---
src/amf/amfd/imm.cc | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/amf/amfd/imm.cc b/src/amf/amfd/imm.cc
index 7b1aa333e..26faffcb5 100644
--- a/src/amf/amfd/
idea to still have this check.
This means that the check if we are on a payload node should be to first check
if rdegetrole exist and if it does then check if the return code is Fail.
Thanks
Lennart
-Original Message-
From: praveen malviya [mailto:praveen.malv...@oracle.com]
Sent: den
Ack.
Thanks
Praveen
On 26-Jun-17 8:35 PM, Lennart Lund wrote:
Fix incorrect detection of node type and misleading information to the user of
ntftest
---
src/ntf/apitest/tet_ntf_clm.c | 14 --
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/src/ntf/apitest/tet_ntf_c
Ack, code review only.
Thanks,
Praveen
On 26-Jun-17 9:25 AM, Minh Chau wrote:
During switchover, standby amfd tries to stop clm tracking,
amfd first got ERR_TIMEOUT and second tries got ERR_NOT_EXIST.
In CLM spec, ERR_TIMEOUT return means the stop clm tracking
may or may not be successful. If
Hi Gary,
Is there any ticket in IMM for any related issue?
Both creation of SU and admin operation on it will go through IMM only.
I do not know how IMM handles it and how IMM spec talks about it.
Does IMM spec allow admin operation on a entity for which all synced
IMMNDS (leaving non synced IMM
Ack.
Thanks
Praveen
On 15-Jun-17 2:01 PM, Hoang Vo wrote:
clmtest 10 3 and 10 4 does not initialize its data struct but reuse
global one lead to failure result when run alone.
---
src/clm/apitest/tet_saClmClusterNotificationFree.c | 12
1 file changed, 12 insertions(+)
diff --g
Ack for the series.
Thanks
PRaveen
On 13-Jun-17 4:54 PM, Hans Nordeback wrote:
---
src/amf/amfnd/avnd_comp.h | 134 ++
src/amf/amfnd/avnd_err.h | 4 +-
src/amf/amfnd/avnd_hc.h | 2 +-
src/amf/amfnd/avnd_proc.h | 10 ++--
src/amf/amfnd/a
Hi Hans,
One comment on this patch inline with [Praveen].
Thanks,
Praveen
On 18-May-17 3:32 PM, Hans Nordeback wrote:
---
src/amf/amfnd/avnd_comp.h | 71 ++---
src/amf/amfnd/avnd_tmr.h | 6 +--
src/amf/amfnd/cam.cc | 2 +-
src/amf/amfnd/chc.cc | 2
Hi Hans,
One comment on this patch inline with [Praveen].
Thanks,
Praveen
On 18-May-17 3:32 PM, Hans Nordeback wrote:
---
src/amf/amfnd/avnd_comp.h | 134 ++
src/amf/amfnd/avnd_err.h | 4 +-
src/amf/amfnd/avnd_hc.h | 2 +-
src/amf/amfnd/av
Ack.
I think this entire if block can be removed and Restarting macro can be
Ored with Assigning macro in if-else block:
diff --git a/src/amf/amfnd/comp.cc b/src/amf/amfnd/comp.cc
index 74b33a3..2f8bb3a 100644
--- a/src/amf/amfnd/comp.cc
+++ b/src/amf/amfnd/comp.cc
@@ -1615,19 +1615,14 @@ uint
I have published v2. immlist will show empty for TIPC.
Thanks,
PRaveen
On 01-Jun-17 7:56 PM, Anders Widell wrote:
Hi!
What is the expected behaviour when using TIPC transport? This is what I
got when I tried:
Name Type Value(s)
=
Ack.
Thanks,
Praveen
On 02-Jun-17 1:01 PM, Hoang Vo wrote:
---
src/clm/apitest/tet_saClmClusterNodeGet.c | 8
src/clm/apitest/tet_saClmClusterTrack.c | 4 ++--
src/clm/apitest/tet_saClmSelectionObjectGet.c | 2 +-
3 files changed, 7 insertions(+), 7 deletions(-)
diff -
ack code review only.
One minor comment:
Magic no 1024 should be replaced by some #define or constant.
Thanks
Praveen
On 01-Jun-17 2:32 PM, A V Mahesh wrote:
---
src/mds/mds_c_api.c | 4 +++-
src/mds/mds_c_db.c| 12 ++--
src/mds/mds_dt_tcp.c | 1 +
src/mds/mds_dt_tipc.c
As per 3.11.1.4.2 Restrictions to Auto-Repair, AMFD has to disable the
SU only when sumaintenance campaign is set.
Ack from me, code review only.
Thanks
Praveen,
On 05-Jun-17 10:35 AM, praveen malviya wrote:
From surestart perspective:
For SURestart recovery AMFD is not informed for
From surestart perspective:
For SURestart recovery AMFD is not informed for disabled state barbecue
it would be spec deviation. While fixing spec deviation for surestart
recovery, I had taken care of this by writing a new function
su_send_suRestart_recovery_msg(). AMFND internally keeps su dis
Ack, code review only.
Thanks,
Praveen
On 11-May-17 3:08 PM, Gary Lee wrote:
> IMM APIs can fail if immnd finishes shutting down before amfd.
> amfd should not unnecessarily assert and cause core dumps
> to be created.
> ---
> src/amf/amfd/app.cc | 7 +--
> src/amf/amfd/comp.cc | 7 +
Ack (not tested).
Thanks
Praveen
On 24-May-17 6:21 AM, Gary Lee wrote:
> Only increment su_cnt_admin_oper for non-opensaf SUs. Otherwise a nodegroup
> admin operation may fail to finish if it is started while a node is joining.
>
> based on analysis from Minh / Praveen
> ---
> src/amf/amfd/sgp
Ack.
Thanks
Praveen
On 19-May-17 12:48 PM, Minh Chau wrote:
> When amfnd-payload responds susi assignment response just before both SC
> go down, and that response message does not come to director. Therefore,
> the status of that assignment could be seen as "modifying" in IMM. When
> SC comes ba
12:41 PM, minh chau wrote:
> Hi Praveen,
>
> Some comments in line with [Minh]
>
> thanks,
> Minh
>
> On 18/05/17 14:54, praveen malviya wrote:
>> Hi Minh,
>>
>> In the description of the ticket there is a log which is :
>> "
>>
7:28.413069 osafamfd [7686:7686:src/amf/amfd/ndfsm.cc:0254]
>> NO Received node_up from 2030f: msg_id 2
>>
>> Looks to me something should not happen inside
>> avnd_evt_mds_avd_dn_evh(). In this avnd_evt_mds_avd_dn_evh(),
>> @is_avd_down should be true, the msg count
avnd_diq_del()) in avnd_di_msg_ack_process().
We need to call avnd_diq_del() in ack message so that msg_id gets updated.
Further looking into it..
Thanks.
Praveen
On 17-May-17 1:50 PM, praveen malviya wrote:
> Hi Minh,
>
> While testing this, I am observing that amfd is dropping the a
Ack.
Thanks
Praveen
On 17-May-17 3:21 AM, Minh Chau wrote:
> When Opensaf 2N switchover, amfd tries to update some attributes
> to IMM. But this time, IMM is not available since Opensaf 2N SI
> is in QUIESCED.
>
> SC-1 osafamfnd[510]: NO Assigned 'safSi=SC-2N,safApp=OpenSAF' QUIESCED to
> 'safS
Hi Minh,
While testing this, I am observing that amfd is dropping the assignment
message because of message id mismatch:
May 17 12:37:39.522117 osafamfd [7686:7686:src/amf/amfd/sgproc.cc:1171]
>> avd_su_si_assign_evh: id:1, node:2030f, act:5,
'safSu=SU1,safSg=AmfDemo,safApp=AmfDemo1', '', ha:3
Hi Minh,
Is this reproducible all the time?
I performed 4 switchvoers on default branch but did not observed.
Thanks
Praveen
On 17-May-17 3:21 AM, Minh Chau wrote:
> When Opensaf 2N switchover, amfd tries to update some attributes
> to IMM. But this time, IMM is not available since Opensaf 2N SI
Hi Long,
This check is very generic.
During su restart cases, a PI SU having NPI components will send
unnecessary enabled events to AMFD. When AMFD will receive this events
it will try to assign this SU and can lead to assignments in other than
2N red models cases.
I think check should be move
Hi Minh,
I am reviewing this patch.
Thanks,
Praveen
On 15-May-17 1:06 PM, Minh Chau wrote:
> When amfnd-payload responds susi assignment response just before both SC
> go down, and that response message does not come to director. Therefore,
> the status of that assignment could be seen as "modif
Hi Long,
I will review it by tomorrow.
Thanks
Praveen
On 15-May-17 8:55 AM, Long Nguyen wrote:
> Dear maintainers,
>
> Can you please help to review the patch?
>
> Thanks so much,
> Long Nguyen.
>
> On 5/9/2017 9:29 AM, Long Nguyen wrote:
>> Hi,
>>
>> Have you had time to look into the patch?
gt; Gary
>
> -----Original Message-
> From: praveen malviya
> Organization: Oracle Corporation
> Date: Wednesday, 26 April 2017 at 7:47 pm
> To: gary
> Cc:
> Subject: Re: [PATCH 1/1] amfd: make auto repair restriction configurable
> [#2435]
>
> Hi Gar
Hi Gary,
If I understand, before 5.2, while running campaign SMF used to set
saAmfSUMaintenanceCampaign attribute in affected SUs using CCB
operations. Since AMF feature "Restrictions to auto repair" was not
implemented (implemented in #2144, 5.2), AMF was still taking actions if
components fa
Ack, code review only.
Thanks,
praveen
On 12-Apr-17 6:06 PM, Nguyen Luu wrote:
> The $piddir variable (containing path to amf_demo comp's pid file)
> is missed to be defined in amf_demo_script.
>
> This could lead to the amf_demo process not getting truely killed
> in some cases when cleanup is
On 11-Apr-17 9:03 PM, Anders Widell wrote:
> A question:
>
> I noticed that clms_send_is_member_info() is called twice from
> clms_imm_node_unlock(), but you only modified the first call (the one
> without PLM). Shouldn't the PLM case also be updated? Is a similar
> change also applicable in
n
Other y
Comments (indicate scope for each "y" above):
-
changeset f4676a9743119edbd642debfe79f3ca43bbb8a47
Author: Praveen Malviya
Date: Wed, 05 Apr 2017 17:56:44 +0530
clm: add tool commands clm-adm, clm-state, clm-fin
opensaf.spec.in |3 +
src/clm/Makefile.am |5 ++
src/clm/tools/clm-adm | 67 ++
src/clm/tools/clm-find | 83 +
src/clm/tools/clm-state | 107
5 files changed
Ack, code review only.
Thanks
Praveen
On 05-Apr-17 11:31 AM, nagendr...@oracle.com wrote:
> src/amf/amfd/su.cc | 3 ++-
> 1 files changed, 2 insertions(+), 1 deletions(-)
>
>
> diff --git a/src/amf/amfd/su.cc b/src/amf/amfd/su.cc
> --- a/src/amf/amfd/su.cc
> +++ b/src/amf/amfd/su.cc
> @@ -1
Ack, code review only.
Thanks
Praveen
On 03-Apr-17 6:53 AM, Gary Lee wrote:
> src/amf/amfd/ndfsm.cc | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
>
> if a duplicate node_up is processed after cluster startup timer is finished,
> don't send a reboot order
>
> diff --git a/src/amf/
src/clm/clmd/clms_evt.c | 2 +-
src/clm/clmd/clms_imm.c | 11 +--
src/clm/clmd/clms_mbcsv.c | 2 +-
src/clm/clmd/clms_util.c | 16 ++--
4 files changed, 17 insertions(+), 14 deletions(-)
In this problem, first user performs CLM lock operation on payload and restart
n
Other n
Comments (indicate scope for each "y" above):
-
changeset 50f1f31babcc78c50b413497d91497266385b997
Author: Praveen Malviya
Date: Mon, 03 Apr 2017 12:47:47 +0530
clms: return TIME_OUT for unloc
src/amf/amfd/clm.cc | 43 ++-
src/amf/amfnd/clm.cc | 6 --
2 files changed, 34 insertions(+), 15 deletions(-)
In reported issue, two CLM nodes are locked simultaneously. For one of the
nodes,
CLM lock gets timed out and user gets REPAIR_PENDING as
n
Other n
Comments (indicate scope for each "y" above):
-
changeset 1ee79821742a117265da9a4d5ba60617ac86e2e4
Author: Praveen Malviya
Date: Mon, 27 Mar 2017 15:25:18 +0530
amf: fix track callback when mul
Hi Mahesh,
5.2RC2 tag is planned on 2017-03-24. I will be reviewing after taggging
on Monday.
Thanks,
Praveen
On 22-Mar-17 1:41 PM, A V Mahesh wrote:
> Hi AMF dev,
>
> A gentle reminder for review.
>
> So far I have received comments for Gary , If you guys doesn't have any
> other comments ,
n
Tests n
Other n
Comments (indicate scope for each "y" above):
-
changeset 9d28b2e0bba4e479bc65c0df6d55d6cc3f71ecd4
Author: Praveen Malviya
Date: Tue, 21 Mar 2017 15:06:48 +0530
amfd: choose CLM unl
src/amf/amfd/clm.cc | 2 ++
src/amf/amfd/sg_2n_fsm.cc | 12 +++-
2 files changed, 13 insertions(+), 1 deletions(-)
When spare controllers are configured in cluster, AMF is chosing CLM locked
controller for
fresh standby controller during failover situation. Currently fresh stan
Thanks
> Gary
>
> -----Original Message-
> From: praveen malviya
> Organization: Oracle Corporation
> Date: Tuesday, 21 March 2017 at 4:49 pm
> To: gary , ,
> ,
> Cc:
> Subject: Re: [PATCH 1 of 1] amfd: remove assignments from lower ranked SU
> while adjusting
Hi Gary,
Thanks for the reviewing the patch.
For readability I had kept separate handling for new definition in the
if block.
Now I will push patch with below refactoring :
/* Check if we need to readjust the SI assignments as
PrefActiveAssignments
got changed
Hi All,
Please review this patch and provide your feedback.
I want to push it on RC2.
Thanks,
Praveen
On 17-Mar-17 2:52 PM, praveen.malv...@oracle.com wrote:
> src/amf/amfd/si.cc | 92
> +++---
> 1 files changed, 46 insertions(+), 46 deletions
n
Tests n
Other n
Comments (indicate scope for each "y" above):
-
changeset 2d69e15801b87ac788b64a8287353616fd7e0c66
Author: Praveen Malviya
Date: Fri, 17 Mar 2017 14:48:12 +0530
amfd: remove assignment
src/amf/amfd/si.cc | 92 +++---
1 files changed, 46 insertions(+), 46 deletions(-)
In N-Way Active mode, when saAmfSIPrefActiveAssignments is reduced,
AMFD removes assignments from higher ranked SU when siranked su is not
configured and lower
ran
e SI assignment is
> not synchronization, I wonder if someone abruptly power off active
> controller when active amfd is about receiving the assignment message,
> or when amfnd just sends out the assignment response message but does
> not reach to amfds?
>
>
>
> On 15/03/17 16:26,
Hi Minh,
Ack with two comments in attached read me.
Thanks,
Praveen
On 10-Mar-17 11:14 AM, minh chau wrote:
Hi all,
Please help to review documentation changes. Files are attached, they
also can be found at below links:
https://sourceforge.net/p/opensaf/tickets/_discuss/thread/342e9c61/8b9a/
Ack, code review only.
Thanks,
Praveen
On 14-Mar-17 11:54 AM, ramesh.bet...@oracle.com wrote:
> src/fm/fmd/fm_main.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
>
> diff --git a/src/fm/fmd/fm_main.c b/src/fm/fmd/fm_main.c
> --- a/src/fm/fmd/fm_main.c
> +++ b/src/fm/fmd/fm_main.
Ack, code review only.
Thanks,
Praveen
On 13-Mar-17 10:24 AM, Nguyen TK Luu wrote:
> src/amf/amfnd/clc.cc | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
>
> Due to a coding error, only the first env variable in the list
> gets printed repeatedly for the length of the list. This ti
src/amf/amfd/sg.cc | 46 +-
src/amf/amfd/sg.h | 1 +
src/amf/amfd/sg_nway_fsm.cc| 38 +++---
src/amf/amfd/sg_nwayact_fsm.cc | 26 ++-
4 files changed, 83 insertions(+), 28 dele
n
Tests n
Other n
Comments (indicate scope for each "y" above):
-
changeset 62b35316b2e40dff6098f4385e2073f2f1e5a11b
Author: Praveen Malviya
Date: Fri, 10 Mar 2017 16:09:20 +0530
amfd: honor PrefAssigne
ents of other levels:
from spec :"The instantiation level is, above all, a means to limit the
load on the system during the instantiation process."
>
> Thanks,
> Minh
>
> On 07/03/17 16:34, praveen malviya wrote:
>> Hi Minh,
>>
>> Is there any harm if both
-Mar-17 4:10 AM, minh chau wrote:
> Hi Praveen,
>
> Please see comments with [Minh5]
>
> Thanks,
> Minh
>
> On 06/03/17 17:52, praveen malviya wrote:
>> Hi Minh,
>>
>> Please see inline with [Praveen].
>>
>> Thanks,
>> Praveen
>>
&
Hi Minh,
Please see inline with [Praveen].
Thanks,
Praveen
On 03-Mar-17 5:39 PM, minh chau wrote:
> Hi Praveen,
>
> I have two comments with [Minh4].
>
> Thanks
> Minh
>
> On 02/03/17 20:49, praveen malviya wrote:
>> Hi Minh,
>> Please see response with
it a/src/amf/amfd/sgproc.cc b/src/amf/amfd/sgproc.cc
>>>> --- a/src/amf/amfd/sgproc.cc
>>>> +++ b/src/amf/amfd/sgproc.cc
>>>> @@ -2092,13 +2092,17 @@ void avd_node_down_mw_susi_failover(AVD_
>>>> * one loop as more than one MW SU per SG
n
Other n
Comments (indicate scope for each "y" above):
-
changeset e3d19c3a42f74d766c8812c4e3713be38c32cee6
Author: Praveen Malviya
Date: Fri, 03 Mar 2017 15:35:11 +0530
clmd: try to re-read node config
src/clm/clmd/clms_imm.c | 97
1 files changed, 73 insertions(+), 24 deletions(-)
While coming up as standby, CLMD successfully initializes with IMM.
It successfuly reads cluster related configuration.
While reading node related configuration from
Ack, code review only.
Thanks,
Praveen.
On 02-Mar-17 3:39 PM, ramesh betham wrote:
> Hi,
>
> Correctionto the patch.
>
> + /* Weird situation in a cluster, where the new-Active
> controller node founds the peer node
> + * (old-Active) is still in the progress of shutdown
t;
>> It seems the component should be re-instantiated if it has no CSI.
>> Whether or not there is an SI assigned should be irrelevant?
>>
>> Thanks
>> Gary
>>
>> -Original Message-
>> From: minh chau
>> Date: Thursday, 23 February 2
Ack with that minor change.
Thanks
Praveen
On 02-Mar-17 1:37 PM, minh chau wrote:
> Hi Praveen,
>
> Two comments in line.
>
> Thanks,
> Minh
>
> On 02/03/17 18:06, praveen malviya wrote:
>> Hi Minh,
>>
>> With the patch opensaf stop is successful but
Ack.
Thanks,
Praveen
On 02-Mar-17 8:17 AM, Tai Dinh wrote:
> src/amf/amfd/nodegroup.cc | 4
> 1 files changed, 4 insertions(+), 0 deletions(-)
>
>
> During instantiation of node group, AMF does not check for the admin state of
> the SG of SUs lead to the result that SUs will be instantiate
Hi,
I am reviewing this patch.
Thanks
Praveen
On 28-Feb-17 8:19 AM, Minh Hon Chau wrote:
> src/amf/amfnd/comp.cc | 13 ++---
> 1 files changed, 10 insertions(+), 3 deletions(-)
>
>
> If node is shutting down, and csi is in assigning state, amfnd deletes cbk
> info which
> is waiting fo
Hi Alex,
Change in comp.cc looks fine.
I guess other changes are related to the node reboot case in the context
of a campaign which performs upgrade by rebooting the node. In this case
since amfnd is freshly coming up, SUs on this node will not have their
sumaintenance attribute set (amfnd do
Hi,
Attached 02_2144.patch is the patch that improves this patch. With
02_2144.patch, AMF will not mark a node disabled when:
-its admin state is locked-in and
-sumaintenance attribute is set for all the sus hosted on this node
including MW.
With this patchg SMF will not get any disabled st
1 - 100 of 1124 matches
Mail list logo
